{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.9813318473112287,
  "eval_steps": 500,
  "global_step": 53500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.03,
      "grad_norm": 1.1059476137161255,
      "learning_rate": 1.981424723692765e-05,
      "loss": 5.2962,
      "step": 500
    },
    {
      "epoch": 0.06,
      "grad_norm": 1.074090838432312,
      "learning_rate": 1.96284944738553e-05,
      "loss": 5.2074,
      "step": 1000
    },
    {
      "epoch": 0.08,
      "grad_norm": 2.9232094287872314,
      "learning_rate": 1.944274171078295e-05,
      "loss": 5.0464,
      "step": 1500
    },
    {
      "epoch": 0.11,
      "grad_norm": 1.1528608798980713,
      "learning_rate": 1.9256988947710598e-05,
      "loss": 4.9315,
      "step": 2000
    },
    {
      "epoch": 0.14,
      "grad_norm": 2.6015784740448,
      "learning_rate": 1.9071236184638247e-05,
      "loss": 4.9604,
      "step": 2500
    },
    {
      "epoch": 0.17,
      "grad_norm": 1.2907854318618774,
      "learning_rate": 1.8885483421565897e-05,
      "loss": 4.825,
      "step": 3000
    },
    {
      "epoch": 0.2,
      "grad_norm": 2.6447720527648926,
      "learning_rate": 1.8699730658493546e-05,
      "loss": 4.6561,
      "step": 3500
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.8539524674415588,
      "learning_rate": 1.8513977895421195e-05,
      "loss": 5.0517,
      "step": 4000
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.9389849901199341,
      "learning_rate": 1.8328225132348844e-05,
      "loss": 4.7113,
      "step": 4500
    },
    {
      "epoch": 0.28,
      "grad_norm": 1.793870210647583,
      "learning_rate": 1.8142472369276493e-05,
      "loss": 4.8306,
      "step": 5000
    },
    {
      "epoch": 0.31,
      "grad_norm": 2.8396799564361572,
      "learning_rate": 1.7956719606204142e-05,
      "loss": 4.6542,
      "step": 5500
    },
    {
      "epoch": 0.33,
      "grad_norm": 1.3629398345947266,
      "learning_rate": 1.7770966843131795e-05,
      "loss": 4.8343,
      "step": 6000
    },
    {
      "epoch": 0.36,
      "grad_norm": 1.432026743888855,
      "learning_rate": 1.7585214080059444e-05,
      "loss": 4.756,
      "step": 6500
    },
    {
      "epoch": 0.39,
      "grad_norm": 1.616821527481079,
      "learning_rate": 1.7399461316987093e-05,
      "loss": 4.5546,
      "step": 7000
    },
    {
      "epoch": 0.42,
      "grad_norm": 1.8740485906600952,
      "learning_rate": 1.7213708553914742e-05,
      "loss": 4.869,
      "step": 7500
    },
    {
      "epoch": 0.45,
      "grad_norm": 3.150221586227417,
      "learning_rate": 1.702795579084239e-05,
      "loss": 4.8451,
      "step": 8000
    },
    {
      "epoch": 0.47,
      "grad_norm": 1.4727258682250977,
      "learning_rate": 1.6842203027770037e-05,
      "loss": 4.8123,
      "step": 8500
    },
    {
      "epoch": 0.5,
      "grad_norm": 1.3298438787460327,
      "learning_rate": 1.6656450264697686e-05,
      "loss": 4.833,
      "step": 9000
    },
    {
      "epoch": 0.53,
      "grad_norm": 1.3734725713729858,
      "learning_rate": 1.647069750162534e-05,
      "loss": 4.2936,
      "step": 9500
    },
    {
      "epoch": 0.56,
      "grad_norm": 1.323432207107544,
      "learning_rate": 1.6284944738552988e-05,
      "loss": 4.0395,
      "step": 10000
    },
    {
      "epoch": 0.59,
      "grad_norm": 1.5172241926193237,
      "learning_rate": 1.6099191975480637e-05,
      "loss": 4.243,
      "step": 10500
    },
    {
      "epoch": 0.61,
      "grad_norm": 1.2672388553619385,
      "learning_rate": 1.5913439212408286e-05,
      "loss": 4.1569,
      "step": 11000
    },
    {
      "epoch": 0.64,
      "grad_norm": 1.1766120195388794,
      "learning_rate": 1.5727686449335935e-05,
      "loss": 4.4089,
      "step": 11500
    },
    {
      "epoch": 0.67,
      "grad_norm": 1.3315163850784302,
      "learning_rate": 1.5541933686263585e-05,
      "loss": 4.8991,
      "step": 12000
    },
    {
      "epoch": 0.7,
      "grad_norm": 1.3195267915725708,
      "learning_rate": 1.5356180923191234e-05,
      "loss": 4.6714,
      "step": 12500
    },
    {
      "epoch": 0.72,
      "grad_norm": 1.1769589185714722,
      "learning_rate": 1.5170428160118885e-05,
      "loss": 4.6833,
      "step": 13000
    },
    {
      "epoch": 0.75,
      "grad_norm": 2.954596996307373,
      "learning_rate": 1.4984675397046534e-05,
      "loss": 4.6754,
      "step": 13500
    },
    {
      "epoch": 0.78,
      "grad_norm": 1.6191766262054443,
      "learning_rate": 1.4798922633974183e-05,
      "loss": 4.5673,
      "step": 14000
    },
    {
      "epoch": 0.81,
      "grad_norm": 1.709163784980774,
      "learning_rate": 1.461316987090183e-05,
      "loss": 4.5477,
      "step": 14500
    },
    {
      "epoch": 0.84,
      "grad_norm": 1.679591417312622,
      "learning_rate": 1.442741710782948e-05,
      "loss": 4.2485,
      "step": 15000
    },
    {
      "epoch": 0.86,
      "grad_norm": 1.4663037061691284,
      "learning_rate": 1.4241664344757129e-05,
      "loss": 4.2211,
      "step": 15500
    },
    {
      "epoch": 0.89,
      "grad_norm": 1.1792720556259155,
      "learning_rate": 1.4055911581684778e-05,
      "loss": 4.1774,
      "step": 16000
    },
    {
      "epoch": 0.92,
      "grad_norm": 2.4920449256896973,
      "learning_rate": 1.3870158818612427e-05,
      "loss": 3.9501,
      "step": 16500
    },
    {
      "epoch": 0.95,
      "grad_norm": 1.2305946350097656,
      "learning_rate": 1.3684406055540078e-05,
      "loss": 4.0126,
      "step": 17000
    },
    {
      "epoch": 0.98,
      "grad_norm": 1.3168740272521973,
      "learning_rate": 1.3498653292467727e-05,
      "loss": 4.4881,
      "step": 17500
    },
    {
      "epoch": 1.0,
      "eval_generated_length": 19.0,
      "eval_loss": 1.565665364265442,
      "eval_rouge1": 0.219,
      "eval_rouge2": 0.1045,
      "eval_rougeL": 0.1858,
      "eval_rougeLsum": 0.1858,
      "eval_runtime": 789.2418,
      "eval_samples_per_second": 14.558,
      "eval_steps_per_second": 0.911,
      "step": 17945
    },
    {
      "epoch": 1.0,
      "grad_norm": 1.1441638469696045,
      "learning_rate": 1.3312900529395376e-05,
      "loss": 4.3781,
      "step": 18000
    },
    {
      "epoch": 1.03,
      "grad_norm": 1.1992547512054443,
      "learning_rate": 1.3127147766323025e-05,
      "loss": 2.8783,
      "step": 18500
    },
    {
      "epoch": 1.06,
      "grad_norm": 1.225577473640442,
      "learning_rate": 1.2941395003250674e-05,
      "loss": 3.0795,
      "step": 19000
    },
    {
      "epoch": 1.09,
      "grad_norm": 1.3939582109451294,
      "learning_rate": 1.2755642240178324e-05,
      "loss": 3.1704,
      "step": 19500
    },
    {
      "epoch": 1.11,
      "grad_norm": 1.366357684135437,
      "learning_rate": 1.2569889477105973e-05,
      "loss": 2.9538,
      "step": 20000
    },
    {
      "epoch": 1.14,
      "grad_norm": 1.182193398475647,
      "learning_rate": 1.2384136714033624e-05,
      "loss": 2.9475,
      "step": 20500
    },
    {
      "epoch": 1.17,
      "grad_norm": 1.2305201292037964,
      "learning_rate": 1.2198383950961273e-05,
      "loss": 2.6189,
      "step": 21000
    },
    {
      "epoch": 1.2,
      "grad_norm": 1.0837234258651733,
      "learning_rate": 1.2012631187888922e-05,
      "loss": 2.6662,
      "step": 21500
    },
    {
      "epoch": 1.23,
      "grad_norm": 1.2700753211975098,
      "learning_rate": 1.1826878424816571e-05,
      "loss": 2.8915,
      "step": 22000
    },
    {
      "epoch": 1.25,
      "grad_norm": 1.0048373937606812,
      "learning_rate": 1.1641125661744218e-05,
      "loss": 2.9367,
      "step": 22500
    },
    {
      "epoch": 1.28,
      "grad_norm": 1.1926288604736328,
      "learning_rate": 1.1455372898671868e-05,
      "loss": 2.9525,
      "step": 23000
    },
    {
      "epoch": 1.31,
      "grad_norm": 2.236363410949707,
      "learning_rate": 1.1269620135599517e-05,
      "loss": 3.1114,
      "step": 23500
    },
    {
      "epoch": 1.34,
      "grad_norm": 1.1560380458831787,
      "learning_rate": 1.1083867372527166e-05,
      "loss": 2.8436,
      "step": 24000
    },
    {
      "epoch": 1.37,
      "grad_norm": 1.6830860376358032,
      "learning_rate": 1.0898114609454817e-05,
      "loss": 2.8528,
      "step": 24500
    },
    {
      "epoch": 1.39,
      "grad_norm": 1.0211920738220215,
      "learning_rate": 1.0712361846382466e-05,
      "loss": 2.8839,
      "step": 25000
    },
    {
      "epoch": 1.42,
      "grad_norm": 1.1330536603927612,
      "learning_rate": 1.0526609083310115e-05,
      "loss": 2.9946,
      "step": 25500
    },
    {
      "epoch": 1.45,
      "grad_norm": 1.6771241426467896,
      "learning_rate": 1.0340856320237764e-05,
      "loss": 3.0686,
      "step": 26000
    },
    {
      "epoch": 1.48,
      "grad_norm": 1.409925103187561,
      "learning_rate": 1.0155103557165413e-05,
      "loss": 3.118,
      "step": 26500
    },
    {
      "epoch": 1.5,
      "grad_norm": 1.1481200456619263,
      "learning_rate": 9.969350794093064e-06,
      "loss": 3.1728,
      "step": 27000
    },
    {
      "epoch": 1.53,
      "grad_norm": 1.0455875396728516,
      "learning_rate": 9.783598031020712e-06,
      "loss": 3.144,
      "step": 27500
    },
    {
      "epoch": 1.56,
      "grad_norm": 1.1866121292114258,
      "learning_rate": 9.59784526794836e-06,
      "loss": 3.0224,
      "step": 28000
    },
    {
      "epoch": 1.59,
      "grad_norm": 1.1943082809448242,
      "learning_rate": 9.41209250487601e-06,
      "loss": 4.6478,
      "step": 28500
    },
    {
      "epoch": 1.62,
      "grad_norm": 1.3221793174743652,
      "learning_rate": 9.22633974180366e-06,
      "loss": 5.313,
      "step": 29000
    },
    {
      "epoch": 1.64,
      "grad_norm": 1.3978910446166992,
      "learning_rate": 9.04058697873131e-06,
      "loss": 5.366,
      "step": 29500
    },
    {
      "epoch": 1.67,
      "grad_norm": 1.257312536239624,
      "learning_rate": 8.854834215658959e-06,
      "loss": 5.4676,
      "step": 30000
    },
    {
      "epoch": 1.7,
      "grad_norm": 1.4381968975067139,
      "learning_rate": 8.669081452586608e-06,
      "loss": 5.3774,
      "step": 30500
    },
    {
      "epoch": 1.73,
      "grad_norm": 2.0183818340301514,
      "learning_rate": 8.483328689514257e-06,
      "loss": 5.2591,
      "step": 31000
    },
    {
      "epoch": 1.76,
      "grad_norm": 1.2171275615692139,
      "learning_rate": 8.297575926441907e-06,
      "loss": 5.4122,
      "step": 31500
    },
    {
      "epoch": 1.78,
      "grad_norm": 1.3053938150405884,
      "learning_rate": 8.111823163369556e-06,
      "loss": 5.3049,
      "step": 32000
    },
    {
      "epoch": 1.81,
      "grad_norm": 1.499539852142334,
      "learning_rate": 7.926070400297205e-06,
      "loss": 5.3328,
      "step": 32500
    },
    {
      "epoch": 1.84,
      "grad_norm": 1.347572684288025,
      "learning_rate": 7.740317637224854e-06,
      "loss": 5.3964,
      "step": 33000
    },
    {
      "epoch": 1.87,
      "grad_norm": 1.2193210124969482,
      "learning_rate": 7.554564874152503e-06,
      "loss": 5.3962,
      "step": 33500
    },
    {
      "epoch": 1.89,
      "grad_norm": 2.145545482635498,
      "learning_rate": 7.368812111080152e-06,
      "loss": 5.2661,
      "step": 34000
    },
    {
      "epoch": 1.92,
      "grad_norm": 1.8138571977615356,
      "learning_rate": 7.183059348007802e-06,
      "loss": 5.3785,
      "step": 34500
    },
    {
      "epoch": 1.95,
      "grad_norm": 1.9416179656982422,
      "learning_rate": 6.997306584935451e-06,
      "loss": 5.3526,
      "step": 35000
    },
    {
      "epoch": 1.98,
      "grad_norm": 1.0953854322433472,
      "learning_rate": 6.8115538218631005e-06,
      "loss": 5.345,
      "step": 35500
    },
    {
      "epoch": 2.0,
      "eval_generated_length": 19.0,
      "eval_loss": 1.5637397766113281,
      "eval_rouge1": 0.2186,
      "eval_rouge2": 0.1038,
      "eval_rougeL": 0.185,
      "eval_rougeLsum": 0.185,
      "eval_runtime": 790.9995,
      "eval_samples_per_second": 14.526,
      "eval_steps_per_second": 0.909,
      "step": 35890
    },
    {
      "epoch": 2.01,
      "grad_norm": 1.6087472438812256,
      "learning_rate": 6.62580105879075e-06,
      "loss": 4.877,
      "step": 36000
    },
    {
      "epoch": 2.03,
      "grad_norm": 1.424822449684143,
      "learning_rate": 6.4400482957184e-06,
      "loss": 3.7964,
      "step": 36500
    },
    {
      "epoch": 2.06,
      "grad_norm": 1.7608376741409302,
      "learning_rate": 6.254295532646049e-06,
      "loss": 3.6988,
      "step": 37000
    },
    {
      "epoch": 2.09,
      "grad_norm": 1.1424431800842285,
      "learning_rate": 6.068542769573698e-06,
      "loss": 3.5234,
      "step": 37500
    },
    {
      "epoch": 2.12,
      "grad_norm": 1.307145595550537,
      "learning_rate": 5.882790006501348e-06,
      "loss": 3.5082,
      "step": 38000
    },
    {
      "epoch": 2.15,
      "grad_norm": 1.7545298337936401,
      "learning_rate": 5.697037243428996e-06,
      "loss": 3.5398,
      "step": 38500
    },
    {
      "epoch": 2.17,
      "grad_norm": 1.009940505027771,
      "learning_rate": 5.5112844803566454e-06,
      "loss": 3.6597,
      "step": 39000
    },
    {
      "epoch": 2.2,
      "grad_norm": 1.2161884307861328,
      "learning_rate": 5.325531717284295e-06,
      "loss": 4.1624,
      "step": 39500
    },
    {
      "epoch": 2.23,
      "grad_norm": 1.7260109186172485,
      "learning_rate": 5.139778954211945e-06,
      "loss": 4.005,
      "step": 40000
    },
    {
      "epoch": 2.26,
      "grad_norm": 1.5710057020187378,
      "learning_rate": 4.954026191139594e-06,
      "loss": 4.0661,
      "step": 40500
    },
    {
      "epoch": 2.28,
      "grad_norm": 1.6778111457824707,
      "learning_rate": 4.768273428067243e-06,
      "loss": 3.9081,
      "step": 41000
    },
    {
      "epoch": 2.31,
      "grad_norm": 1.3847213983535767,
      "learning_rate": 4.582520664994892e-06,
      "loss": 3.8309,
      "step": 41500
    },
    {
      "epoch": 2.34,
      "grad_norm": 1.128096342086792,
      "learning_rate": 4.396767901922541e-06,
      "loss": 3.6819,
      "step": 42000
    },
    {
      "epoch": 2.37,
      "grad_norm": 1.5037422180175781,
      "learning_rate": 4.21101513885019e-06,
      "loss": 3.566,
      "step": 42500
    },
    {
      "epoch": 2.4,
      "grad_norm": 1.3365986347198486,
      "learning_rate": 4.02526237577784e-06,
      "loss": 3.2959,
      "step": 43000
    },
    {
      "epoch": 2.42,
      "grad_norm": 1.167784333229065,
      "learning_rate": 3.8395096127054895e-06,
      "loss": 3.4172,
      "step": 43500
    },
    {
      "epoch": 2.45,
      "grad_norm": 1.3119831085205078,
      "learning_rate": 3.6537568496331386e-06,
      "loss": 3.5034,
      "step": 44000
    },
    {
      "epoch": 2.48,
      "grad_norm": 1.2421696186065674,
      "learning_rate": 3.4680040865607878e-06,
      "loss": 4.1109,
      "step": 44500
    },
    {
      "epoch": 2.51,
      "grad_norm": 1.1798334121704102,
      "learning_rate": 3.2822513234884373e-06,
      "loss": 4.0948,
      "step": 45000
    },
    {
      "epoch": 2.54,
      "grad_norm": 1.1406497955322266,
      "learning_rate": 3.0964985604160865e-06,
      "loss": 3.9285,
      "step": 45500
    },
    {
      "epoch": 2.56,
      "grad_norm": 1.6897916793823242,
      "learning_rate": 2.9107457973437356e-06,
      "loss": 3.705,
      "step": 46000
    },
    {
      "epoch": 2.59,
      "grad_norm": 1.162559151649475,
      "learning_rate": 2.724993034271385e-06,
      "loss": 3.4343,
      "step": 46500
    },
    {
      "epoch": 2.62,
      "grad_norm": 1.2140285968780518,
      "learning_rate": 2.5392402711990344e-06,
      "loss": 3.2604,
      "step": 47000
    },
    {
      "epoch": 2.65,
      "grad_norm": 1.20240318775177,
      "learning_rate": 2.3534875081266835e-06,
      "loss": 3.5073,
      "step": 47500
    },
    {
      "epoch": 2.67,
      "grad_norm": 1.1845455169677734,
      "learning_rate": 2.167734745054333e-06,
      "loss": 3.1102,
      "step": 48000
    },
    {
      "epoch": 2.7,
      "grad_norm": 1.2270458936691284,
      "learning_rate": 1.9819819819819822e-06,
      "loss": 3.1802,
      "step": 48500
    },
    {
      "epoch": 2.73,
      "grad_norm": 1.1969292163848877,
      "learning_rate": 1.7962292189096314e-06,
      "loss": 3.1325,
      "step": 49000
    },
    {
      "epoch": 2.76,
      "grad_norm": 1.2453925609588623,
      "learning_rate": 1.6104764558372807e-06,
      "loss": 2.9085,
      "step": 49500
    },
    {
      "epoch": 2.79,
      "grad_norm": 1.3069303035736084,
      "learning_rate": 1.42472369276493e-06,
      "loss": 2.9761,
      "step": 50000
    },
    {
      "epoch": 2.81,
      "grad_norm": 1.4485862255096436,
      "learning_rate": 1.2389709296925793e-06,
      "loss": 3.5526,
      "step": 50500
    },
    {
      "epoch": 2.84,
      "grad_norm": 1.4915004968643188,
      "learning_rate": 1.0532181666202286e-06,
      "loss": 3.7851,
      "step": 51000
    },
    {
      "epoch": 2.87,
      "grad_norm": 1.271291971206665,
      "learning_rate": 8.674654035478779e-07,
      "loss": 3.9262,
      "step": 51500
    },
    {
      "epoch": 2.9,
      "grad_norm": 1.3475127220153809,
      "learning_rate": 6.817126404755271e-07,
      "loss": 4.0264,
      "step": 52000
    },
    {
      "epoch": 2.93,
      "grad_norm": 1.2061536312103271,
      "learning_rate": 4.959598774031764e-07,
      "loss": 3.9977,
      "step": 52500
    },
    {
      "epoch": 2.95,
      "grad_norm": 1.8415242433547974,
      "learning_rate": 3.102071143308257e-07,
      "loss": 4.0098,
      "step": 53000
    },
    {
      "epoch": 2.98,
      "grad_norm": 1.0884873867034912,
      "learning_rate": 1.2445435125847498e-07,
      "loss": 4.0167,
      "step": 53500
    }
  ],
  "logging_steps": 500,
  "max_steps": 53835,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 2.317005244339323e+17,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}